Anything new about paternal contribution to reproductive outcomes? :

(1)

INTRODUCTION

Paternal health and behavioral lifestyles affect maternal and neonatal outcomes but remain neglected topics in reproductive health. Even though their impor-

tance has been globally acknowledged [1], health care services in Europe have failed so far to attract and increase the involvement of fathers-to-be in reproductive programs [2].

Health campaigns used to focus only on the need to

Received: Aug 4, 2020 Revised: Oct 13, 2020 Accepted: Nov 8, 2020 Published online Jan 8, 2021 Correspondence to: Alberto E. Tozzi https://orcid.org/0000-0002-6884-984X

Predictive and Preventive Medicine Research Unit, Bambino Gesù Children’s Hospital, IRCCS, Piazza di Sant'Onofrio, 4, 00165 Rome, Italy.

Anything New About Paternal Contribution to

Reproductive Outcomes? A Review of The Evidence

Caterina Montagnoli^1,2 , Stefania Ruggeri³ , Giulia Cinelli⁴ , Alberto E. Tozzi⁴ , Chiara Bovo¹ , Renata Bortolus⁵ , Giovanni Zanconato⁶

1Department of Medical Direction, Verona University Hospital, Verona, Italy, ²Department of Midwifery, Geneva School of Health Sciences, HES-SO University of Applied Sciences and Arts of Western Switzerland, Geneva, Switzerland, ³Research Centre for Food and Nutrition- CREA, ⁴Predictive and Preventive Medicine Research Unit, Bambino Gesù Children’s Hospital, IRCCS, ⁵Directorate General for Preventive Health - Office 9, Ministry of Health, Rome, ⁶Department of Surgery, Odontostomatology and Maternal and Child Health, University of Verona, Verona, Italy

Paternal health and behavioral lifestyles affect reproductive and neonatal outcomes and yet the magnitude of these effects remain underestimated. Even though these impacts have been formally recognized as a central aspect of reproductive health, health care services in Europe often neglect the involvement of fathers in their reproductive programs. Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for systematic reviews, a literature search was carried out to assess the possible impact of paternal health on reproductive outcomes. The comprehensive strat- egy included cohort studies and meta-analysis available on PubMed, Web of Science, CINAHL, and Google scholar. Cross- referencing of bibliographies of the selected papers ensured wider study capture. Paternal factors were grouped into two categories respectively identified with the terms “Biological Paternal Factors” and “Lifestyle Paternal Factors”. Advanced age may impair male fertility and affect early pregnancy stages. Increased body mass index, smoking, alcohol and recreational drugs, all alter seminal fluid parameters. Hazardous alcohol use correlates with low birthweight in pregnancy and harmful behavioral lifestyles have been linked to congenital heart defects, metabolic and neurodevelopmental disorders in the offspring.

Measures targeting paternal health and lifestyle within the first 1,000 days’ timeframe need to be implemented in couples undergoing reproductive decisions. Health professionals, as well as future fathers, must be aware of the benefits for the offspring associated with correct paternal behaviors. More research is needed to build guidelines and to implement specific programs aiming at reproductive health promotion.

Keywords:

Keywords: Fertility; Life style; Long term adverse effects; Maternal health; Paternal exposure

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

pISSN: 2287-4208 / eISSN: 2287-4690

World J Mens Health Published online Jan 8, 2021 https://doi.org/10.5534/wjmh.200147

Health promotion, disease prevention, and lifestyle

(2)

improve maternal health, relegating to a secondary role and only marginally involving fathers in labor and childbirth.

Women are regularly encouraged to take care of their health and monitor lifestyle habits, whereas few specific recommendations concern the male partner [3].

In spite of this prejudice, it has become evident that the male factor is relevant and affects fertility as well as pregnancy outcomes. Medical research has shown that alongside their supportive role, men can improve perinatal outcomes by optimizing their own health and behavioral lifestyles [4,5].

Although the magnitude of male contribution is still undetermined, recent studies provide evidence that justifies the introduction of effective paternal health measures such as genetic factors screening in light of the possible association with recurrent pregnancy loss [4,6,7].

Previous reviews recognize paternal age as a possible risk factor for preterm birth (PTB), genetic abnormali- ties, cancer development, and other musculoskeletal congenital syndromes in childhood [8]. Considering the existing gap of available studies exploring paternal and maternal health, likely the result of traditional gender stereotypes and a “macho” attitude towards reproductive matters [9], this review aims to draw attention to paternal influence upon the descendants’ health.

The broader approach adopted in our paper provides a more comprehensive overview considering every potentially relevant factor related to paternal biological and behavioral lifestyles, as opposed to existing reviews which generally focus on single paternal aspects.

MAIN BODY

1. Materials and methods

1) Data sources and searches

All human studies published up to 2020 reporting on paternal exposure factors and lifestyles associated with reproductive outcomes were identified using PubMed, CINHAL, and Web of Science. Cross-referencing in bibliographies of the appraised papers ensured wider study capture.

Our initial search was not limited to any particular type of study and all potentially eligible studies were reviewed. The electronic search encompassed keywords referring to the periconceptional time period, us-

ing the following key concepts and related keywords:

“body mass index BMI”, “alcohol consumption”, “smoking habits”, “medical comorbidities and therapeutic treatments”, “occupational hazards”, “environmental hazards”, “recreational or illicit drug use”, “paternal advanced age”, and “reproductive outcomes”. Results were categorized in terms of early periconceptional morbidities (e.g., infertility) and later pregnancy outcomes inclusive of congenital anomalies (CAs), small for gestational age, low birthweight (LBW) babies, and PTBs.

2) Study selection

Studies which assessed associations between paternal health condition/habits and maternal/fetal/neonatal complications, or reproductive outcomes were reviewed.

Inclusion criteria for selected papers were years (from inception of 2008 until March 2020), study type (case- control, cohort, randomized controlled trialss, and meta- analysis), availability of full text, humans as subjects, and English as language of publication. Animal studies, case reports and review articles were excluded.

3) Data extraction and quality assessment

The full text of eligible papers was obtained, and quality was assessed according to the number of subjects involved and the statistical significance of the results presented in each study. Studies providing adjusted odds ratios (aOR), 95% confidence intervals (CI) and p-value were favored. Description of the studied population was also valued as a quality index as well as control for maternal effects in the paternal model.

Adjustment habitually accounted for maternal socio- economic and biological information when performing paternal data analysis [10]. PRISMA (Preferred Report- ing Items for Systematic Reviews and Meta-Analyses) Statements tools, i.e., “Checklist for Systematic reviews and Meta-Analysis items” and “Flow chart template”

[11], were used to build the flow chart presented in the Results section and to checklist essential items of the current review.

2. Results

A total of 520 articles were identified. Cross-referencing in bibliographies of the initial selected papers added 16 additional studies. Three of them, despite being published before 2008, were included due to their relevance to this topic. After both title and abstract

(3)

screening, 94 papers were assessed for eligibility. Thir- ty-nine studies were excluded, either because they did not meet the criteria for the specific topic searched, or because full text was not available. In the end, 53 articles were included, read and analyzed.

In the following search-flow diagram (Fig. 1) the whole process of identification, screening, eligibility, and inclusion is graphically summarized.

Selected paternal indicators identify two distinct categories respectively labelled with the terms “Biologi- cal Paternal Factors” and “Lifestyle Paternal Factors”.

The former includes variables, such as age, body mass index (BMI), medical comorbidities, and related therapies. The latter comprises paternal exposure to external occupational hazards or acquired health determinants, such as smoking habits, alcohol consumption, and recreational drug use. For each group three kinds of reproductive outcomes were searched: male fertility, early pregnancy complications, fetal and postnatal outcomes.

Results were organized into tables which included the following items: Author’s name and year of publication; study design and study period or research timeframe for literature reviews; number of participants/

studies; type of exposure factor; specific reproductive outcome; statistical strength of association mainly expressed in OR and CIs; key findings. Additional notes

for a better understanding (e.g., abbreviations and com- ments on results) are at the end of each table.

1) Biological paternal factors

Biological factors related to male partner conditions and their effects upon reproductive outcomes find limited space in the medical literature. Nevertheless, paternal aging, body weight and overall health status (i.e., medical comorbidities and related therapies) are indicators studied by different authors whose findings provide enough information to draw initial conclusions (Table 1, 2).

(1) Fertility

Parental aging and declining fertility are generally considered to be correlated. Resembling what happens in women, paternal age could reduce the chance of conceiving in terms of seminal fluid parameters as well as potential CAs. The latest literature confirms that alterations of sperm characteristics include reduced semen volume and teratozoospermia [12]. However evaluation of sperm DNA might have greater clinical utility since seminal DNA fragmentation index (DFI) is found to be higher in men aged 40 or older [12] and this might explain the demographic trend of a steady decrease of men’s fertility after the age of 39 [13].

The impact of paternal BMI upon fertility has drawn

Fig. 1. Flow chart of study-selection pro- cess.

IncludedEligibilityScreeningIdentification

Records screened (n=536)

Full-text articles assessed for eligibility

(n=94)

Studies included in qualitative synthesis

(n=55)

Records excluded after title and abstract screening

(n=442) Records identified through

database searching (n=520)

Additional records identified through other sources

(n=16)

Full-text articles excluded, with reasons

(n=39)

(4)

Table 1. Biological paternal factors and fertility CitationStudy designSubjectPaternal factorReproductive outcomeOR (95% CI), p-value, overall trendKey finding Rosiak-Gill et al (2019) [12]Cohort1,124 malesAge≥40 yDFIa >10%OR 2.31 (1.36–3.92) p<0.01Detrimental effect of advanced paternal age on sperm chromatin integrity. Higher incidence of >10% DFI in subjects aged ≥40 years. Matorras et al (2011) [13]Cohort454,753 birthsAgeFertility21%–23% yearly reductionMale fertility starts decreasing at 39 years of age. Capelouto et al (2018) [17]Cohort949 female recipi- entsAge≥51 yImplantation rate Clinical pregnancy rate Live birth rate aRR 0.95 (0.71–1.27) aRR 1.03 (0.82–1.29) aRR 1.03 (0.75–1.41)

After adjusting for variables known to affect oocyte donation cycle success, paternal age and BMI were not associated with differences in IVF outcomes. BMI≥35 kg/m2 aRR 0.93 (0.69–1.24) aRR 0.92 (0.72–1.19) aRR 1.07 (0.81–1.42) Nguyen et al (2007) [14]Cohort26,303 pregnanciesBMI 25–29.9 kg/m2 InfertilityOR 1.20 (1.04–1.38)Direct correlation between BMI and infertility indicators (altered T, E2 levels, and poor semen quality).BMI 30–34.9 kg/m2 OR 1.36 (1.13–1.63) Tunc et al (2011) [15]Cohort81 malesBMISeminal oxidative stressr=0.23 p<0.05Positive correlation between BMI and seminal oxidative stress. An inverse correlation was found between BMI and sperm concentration. Small sample results have limited statistical value.Sperm concentrationr=-0.33 p<0.01 Kort et al (2006) [16]Cohort520 malesBMI 18–25 kg/m2 BMI 25–29.9 kg/m2 BMI 30–34.9 kg/m2

DFI19.9%±1.96% 25.8%±2.23% 27.0%±3.16%

As paternal BMI increases, chromatin‐intact normal‐ motile sperm cells per ejaculate decrease. BMIp<0.05 Eisenberg et al (2016) [18]Cohort41 malesDiabetesFecundityOR 0.35 (0.13–0.88)Paternal diabetes diminishes a couple’s fecundity expressed by longer TTP. Evans-Hoeker et al (2018) [22]Cohort1,608 malesMDFecundityRR 0.44 (0.20–0.98)Female partners of currently depressed males (n=34) are less likely to achieve conception. Tanrikut et al (2010) [23]Cohort35 malesAntidepressantsDNA fragmentationOR 9.33 (2.3–37.9)The fertility potential of men on paroxetine may be adversely affected by sperm DNA increased fragmentation. Bourke et al (2019) [21]Cohort205 malesCystic fibrosisNatural conception0.5%Men with cystic fibrosis are nearly always infertile. As- sisted reproduction gives good chances for biological parenthood in males reaching adult age.ART conception9% Roessner et al (2012) [19]Case-control27 diabetics males and 18 healthy controls

Healthy casesDFI8.2%±5.4%ROS and DNA fragmentation affecting sperm quality are increased in both types of diabetes. This effect is more pronounced in men with diabetes type II. Diabetes type IDFI22.5%±8.4% p<0.01 Diabetes type IIDFI28.9%±18.4% p<0.01 ROS19.7%±23.1% p<0.05

(5)

Table 1. Continued CitationStudy designSubjectPaternal factorReproductive outcomeOR (95% CI), p-value, overall trendKey finding Ananthakrishnan et al (2019) [25]Cohort256 malesIBDReduced fecundityOR 1.99 (1.05–3.78)IBD likely causes delay in conception. Amino salicylate used to control IBD has no significant effect on fecundity.Amino salicylate usep=0.09 Hayashi et al (2008) [24]Case-control73 intervention and 92 control groupH1 receptor antagonists, antiepileptics and antibiotics

Semen quality93% vs. 12%Long-term use of drugs can induce fertility hazard in terms of asthenozoospermia and oligozoospermia.Conception rate85% vs. 10% OR: odds ratio, CI: confidence interval, DFI: DNA fragmentation index, BMI: body mass index, aRR: adjusted risk ratio, IVF: in-vitro fertilization, T: testosterone, E2: rstradiol, TTP: time to pregnancy, MD: major depression, RR: risk ratio, ART: artificial reproductive techniques, ROS: reactive oxygen species, IBD: inflammatory bowel disease. a DFI is an expression of sperm quantity and quality. Table 2. Biological paternal factors and pregnancy outcomes CitationStudy designSubject Paternal factorReproductive outcomeOR (95% CI), p-value, overall trendKey finding Hurley and DeFranco (2017) [27]

Cohort1,034,552 birthsAgePreeclampsiaRR 1.00 (0.99–1.00)Paternal aging is not associated with preeclamp- sia nor any selected neonatal outcomes after accounting for maternal age.PTBRR 1.01 (1.01–1.01) Fetal growth restrictionRR 1.01 (1.01–1.02) CAsRR 1.01 (0.99–1.03) Genetic disorderRR 1.01 (0.99–1.04) NICU admissionRR 1.01 (1.01–1.01) Tamura et al (2018) [26]Cohort18,059 mother–infant dyadsAgePTBRR 1.22 (1.05–1.42) p<0.01Compared to newborns with younger fathers (25– 34 years), rates of PTB and VLBW among infants with older fathers (≥35 years) are significantly higher. Results adjusted for maternal confound- ing factors, such as ART, BMI, PTB, SGA, VLBW.

VLBWRR 2.02 (1.22–3.35) p<0.01 Magnus et al (2018) [30]Cohort132,331 infantsBMI>30 kg/m2 Type I diabetesaHR 1.51 (1.11–2.04)Childhood-onset type I diabetes possibly influ- enced by paternal obesity. Noor et al (2019) [28]Cohort429 father-mother- infant triadsBMI≥25 kg/m2 Increased birthweightp<0.01Higher paternal BMI (≥25 kg/m2 ) is associated with increased offspring birthweight.

(6)

Table 2. Continued 1 CitationStudy designSubject Paternal factorReproductive outcomeOR (95% CI), p-value, overall trendKey finding Mei et al (2018) [29]Cohort1,178 live birthsBMI≥25 kg/m2 Birthweightβ=0.04 p>0.05Paternal overweight/obesity has no effect on BMI_Z score at birth. Significant but mild paternal influence only detected after birth. Moss and Harris (2015) [32]Cohort372 infantsDiabetesLBWp<0.01Offspring’s lower birthweight associated with paternal diabetes. Δ weight=-783.9 g (range -1,014.2 to -553.6). Ji et al (2018) [33]Cohort15,615 malesType I diabetesADHDaHR 1.20 (1.03–1.41)A paternal history of type I diabetes is associated with a 20% increase in risk of being diagnosed with ADHD. Lydholm et al (2019) [89]Cohort1,206,600 infantsInfection/anti-infective agentsMental disorder in the offspringHR 1.01 (0.98–1.03)No paternal association with an increased risk of mental disorders in the child. Buck Louis et al (2018) [35]Cohort4,886 pregnanciesHistory of mood/anxi- etyPonderal indexa β 0.05 (0.001–0.09)Positive association between paternal history of mood/anxiety and ponderal index. Autoimmune disease was associated with larger head circumference.Autoimmune diseaseHead circumferenceβ 0.87 (0.15–1.60) Midtvedt et al (2017) [37]Cohort230 fathers and 350 childrenMycophenolic acidCAs3.9% vs. 2.6% p=0.49Mycophenolic acid is the active immunosuppres- sive substance used after organ transplantation. No significant association with risk of malformations in the offspring. Birthweight is also similar in exposed and unexposed cohorts of children.

Birthweight3,381±681 g vs. 3,429±714 g p=0.53 Lopez-Lopez et al (2018) [38]

Cohort33 fathers and 49 childrenMycophenolic acidCAsNo casesNo increased incidence of malformations in descendants of male kidney transplantation recipients. Small sample. Viktorin et al (2018) [36]Cohort3,983 cases and 164,492 controlsAntidepressantsPTBaOR 0.91 (0.79–1.04)Paternal intake of antidepressants around conception is safe with respect to preterm birth, malformations, autism, and intellectual disabilities. CAsaOR 1.06 (0.90–1.26) Intellectual disabilityb aOR 0.82 (0.51–1.31) Autism spectrum disorderaHR 1.13 (0.84–1.53) Yang et al (2018) [34]Cohort781,470 singletonsAntidepressants SSRIADHDHR 1.26 (1.06–1.51)The increased risk of ADHD in offspring associated with paternal SSRI used 12 to 3 months before conception, could be due to the underlying paternal conditions related to drug use. Ananthakrish- nan et al (2019) [25]

Cohort256 malesBiologics CAsOR 0.50 (0.05–5.20)Biologics, thiopurine, corticosteroid use to control IBD has no significant effect on CAs.Thiopurine OR 0.49 (0.05–4.82) CorticosteroidsOR 1.80 (0.30–10.94) Yang et al (2019) [40]Cohort733,282 singletonsAntiepileptic drugs useCAsOR 1.23 (1.10–1.37)The increased risk of CAs may be attributable to the underlying conditions requiring antiepilep- tic drugs use.a. Preconceptional b. In pregnancyOR 1.29 (1.03–1.61) OR 1.35 (1.12–1.65)

(7)

the attention of researchers: body weight affects hor- monal balance and semen quality, the latter undergoing deterioration as body mass increases [14]. Seminal oxidative stress, DFI, abnormal sperm concentration, and motility are often observed in men with higher BMI [15,16]. Given the widespread use of assisted reproductive technologies (ART) to counteract the effects of age on chances of natural conception, recent data are reassuring: in-vitro fertilization (IVF) outcomes in terms of implantation, clinical pregnancy and live birth rates do not seem to be affected by paternal age and BMI [17].

Paternal concurrent diseases and related therapies are a matter of concern when evaluating male fertility. Diabetes, a high prevalence metabolic disease, has been associated with reduced fecundity, longer time to pregnancy (TTP) rates and also with higher reactive oxygen species (ROS) and sperm DNA fragmentation [18], which raises implications for fetal development [19].

Parenting and reproductive health are a concern for individuals affected by chronic conditions. Although a rare disease, estimated to affect approximately 1 in 3,000 individuals, cystic fibrosis (CF) has severe implications on male fertility [20]. The increasing life expec- tancy of the CF population, with more frequent tran- sition to adulthood, has raised the number of young males wishing to become parents. Documented cases of CF men conceiving naturally are estimated less than 1%, while biological fatherhood may be achieved in 9%

to 10% of subjects undergoing ART sperm retrieval [21].

Paternal mental health is relevant when it comes to fecundity: female partners of depressed males are less likely to achieve conception [22]. This finding could depend upon use of antidepressants that have been linked to higher sperm DNA fragmentation and reduction of the couple’s fertility [23].

Medical treatments may negatively affect male fertility via different mechanisms. Although evidence suggests that sexual function is reversible after thera- py discontinuation, treatment suspension is not always an option for patients with chronic conditions. In a Japanese study, semen of patients treated with antiepileptics and H1 receptor antagonists to cure asthma or chronic bronchitis more often show oligozoospermia and asthenozoospermia [24]. The use of aminosalicylate to control inflammatory bowel disease has been associated with longer TTP [25]. These results suggest that some drugs, used on a long-term basis, cause fertil- Table 2. Continued 2 CitationStudy designSubject Paternal factorReproductive outcomeOR (95% CI), p-value, overall trendKey finding Larsen et al (2016) [31]CohortNewborns 372 cases and 399,498 controls

Anti-TNF-α agents’ treatmentPTBOR 1.42 (0.52–3.86)No association between reproductive outcomes and paternal preconceptional exposure to anti-TNF-α agents in patients with inflammatory bowel, rheumatologic or dermatological diseases.

SGAOR 1.70 (0.74–3.89) Wallenius et al (2015) [39]CohortNewborns 110 cases and 230 controlsDMARDs treatmentCAsRR 1.22 (0.45–3.31) p=0.69Preconceptional paternal exposure to DMARDs, whether synthetic or biological, does not increase CAs. OR: odds ratio, CI: confidence interval, PTB: preterm birth, CAs: congenital anomalies, NICU: neonatal intensive care unit, RR: risk ratio, VLBW: very low birthweight, ART: artificial reproductive techniques, BMI: body mass index, SGA: small for gestational age, aHR: adjusted hazard ratio, LBW: low birthweight, ADHD: attention-deficit/hyperactivity disorder, HR: hazard ratio, aOR: adjusted odds ratio, SSRI: selective serotonin reuptake inhibitors, IBD: inflammatory bowel disease, TNF: tumor necrosis factor, DMARDs: disease-modifying antirheumatic drugs. a Ponderal index=[birthweight (g)/length (cm3 )]×100. It assesses infant adiposity. b As defined by ICD (International Classification of Diseases)-10.

(8)

ity hazards reversible only when alternative medical treatment is available.

(2) Pregnancy outcomes

Available studies report conflicting results with regard to the implications of age on preterm delivery and birthweight. In a large Japanese cohort study, controlled for maternal age, pregnancies fathered by men of 35 years of age or older end more often prema- turely (p<0.01) and with very low birthweight (VLBW) babies (p<0.01) [26]. These findings are not confirmed by a North American study which reports no association between age, increased risk of PTB, fetal growth restriction [27]. Multiple different factors in the two settings, i.e., standard diets, climate, and quality of life, could explain the contrasting conclusions.

Unlike age, paternal excessive BMI seems to cor- relate with higher birthweight increasing the risks of obesity and type I diabetes in childhood [28-30].

Men affected by health conditions are likely to ex- pose their progeny to developmental disorders: paternal diabetes and inflammatory bowel disease increase the risk of reduced fetal weight gain [31-33], while history of mood or anxiety disorders, as well as the use of antidepressants at the time of conception, may also lead to sequelae in the offspring ranging from development of attention-deficit/hyperactivity disorders (ADHDs) to increased infant ponderal index [34,35]. Against this background, findings of a Swedish cohort are reassuring since no correlation is observed between paternal antidepressants use and intellectual disabilities, as defined by ICD (International Classification of Dis- eases)-10 [36].

Generally, no other paternal medical treatments have been correlated to impaired fetal structural development and growth. Specifically, there is no evidence that fetal damage may be caused by antiepileptics, thiopurine and corticosteroid drugs nor disease- modifying antirheumatic drugs or mycophenolic acid [25,31,34,37-40].

2) Lifestyle paternal factors

A whole set of factors related to environment, oc- cupation and habits is known to exert effects of vari- able severity upon reproductive outcomes. A prolonged exposure may permanently impair paternal sexual organs’ function and have long-term negative implications on the offspring. Table 3, 4 summarize the latest _{Table 3.}

External paternal factors and fertility CitationStudy designSubjectExposureReproductive outcomeOR (95% CI), p-value, overall trendKey finding Tang et al (2019) [41]Cohort1,631 malesSmokingSemen quality, semen volume, oligospermia, sperm motilityp<0.001 p<0.05 p<0.05

Cigarette smoking (≥10 packs/y) is associated with lower semen volume and total sperm count but increased motility. Rehman et al (2019) [42]Cross-section- al165 infertile vs. 211 fertile malesSmokingT and SHBG levels, semen qualityp<0.05Reduced sexual hormone levels which alters sperm total count and morphology but do not impact their motility. Gaur et al (2010) [43]Cohort100 cases vs. 100 controlsSmokingAsthenozoospermiap<0.0001Positive correlation between quantity of cigarettes smoked or amount of alcohol intake and altered sperm parameters but alterations are observed even at low degrees of alcohol/tobacco addiction.

Teratozoospermiap<0.05 100 cases vs. 100 controlsAlcoholOligozoospermiap<0.05 Teratozoospermiap<0.001 Borges et al (2018) [44]Cohort965 malesSmokingSemen quality, fertilization and blastocyst formation ratep<0.05Cigarette smoking and alcohol consumption reduce semen quality, in terms of total sperm count, fertilization rates, and blastocyst formation rates.AlcoholDNA fragmentationp<0.01 Sperm concentration, fertilization and blastocyst formation ratep<0.05

(9)

Table 3. Continued CitationStudy designSubjectExposureReproductive outcomeOR (95% CI), p-value, overall trendKey finding Kasman et al (2018) [46]Cohort758 males and 1,076 femalesMarijuana TTPTR 1.08 (0.79–1.47)No significant impact of marijuana use on TR. Wise et al (2017) [47]Cohort1,125 couplesMarijuana use <1 time/wk ≥1 time/wk

FecundabilityFRs 0.87 (0.66–1.15) FRs 1.24 (0.90–1.70)FRs indicate no association between marijuana use and fecundability. Nassan et al (2019) [48]Cohort662 sub-fertile malesMarijuana use current/past never

Semen quality, concentration and motilityp<0.001After adjusting for potential confounders, marijuana users have significantly higher sperm concentration and motility than never users. Gundersen et al (2015) [100]

Cohort1,215 malesMarijuanaSperm concentration1.07 (0.62–1.87)After adjustment for confounders, regular marijuana smoking was not associated with lower sperm concentration nor total sperm count.Sperm total count1.14 (0.69–1.89) Pichini et al (2012) [45]Cohort164 couples and 24 individualsCannabinoids and cocaine users vs. non-users

Urinary Tp<0.05Male drug consumption was associated with significantly lower urinary T concentrations (mean 105.5±21.1 SD vs. 124.5±46.5 mg/24 h). Tielemans et al (2000) [51]Cohort726 couplesOrganic solventsIVF implantation rateOR 0.24 (0.06–0.91)Couples with male partner exposed to organic solvents have reduced IVF implantation rates. Level of exposure matters. No reduction found after exposure to pesticides, metal dust and fumes or welding fumes.

PesticidesOR 1.57 (0.33–7.44) Metal dust and fumesOR 1.44 (0.57–3.61) Welding fumesOR 1.20 (0.31–4.68) Dodge et al (2015) [52]Cohort218 couplesMethyl parabenLive birth rate after IUIaOR 0.19 (0.04–0.82)Paternal professional exposure to methyl paraben at a specific concentration (10.5~29.0 ng/mL) is associated with decreased odds of live birth following IUI. Campagna et al (2015) [53]Cohort1,223 couplesDDTEarly fecundityFR 1.22 (0.84–1.77)Among the spouses of DDT workers (the anti- malaria environmental agent), fecundability did not vary during DDT handling nor in the following decade.

Fecundity 10 years after exposureFR 1.01 (0.67–1.50) Buck Louis et al (2016) [50]Cohort501 couplesHeavy metals/non- persistent chemicals Fecundability and specific exposure Lead benzophenone-2, mBz-phthalate, mM-phthalate OR 0.83 (0.70–0.98) OR 0.69 (0.49–0.97) OR 0.80 (0.67–0.97) OR 0.81 (0.70–0.94) Specific exposure within persistent metals/chemicals and non-persistent chemicals is associated with a significant reduction in couples’ fecundability. OR: odds ratio, CI: confidence interval, T: testosterone, SHBG: sex hormone binding globulin, TTP: time to pregnancy, TR: TTP ratio, FR: fecundability ratio, SD: standard deviation, IVF: in-vitro fertilization, aOR: adjusted odds ratio, IUI: intra-uterine insemination, DDT: dichloro-diphenyl-trichloroethane.

(10)

Table 4. External paternal factors and pregnancy outcomes CitationStudy designSubjectExposureLong-term reproductive outcomeOR (95% CI), p-value, overall trendKey finding Magnus et al (2019) [55]Cohort1,381 cases and 618,322 controls

SmokingDiabetesHR 0.97 (0.82–1.2)No association between paternal or second-hand smoking during pregnancy with childhood-onset type I diabetes. Mejia-Lancheros et al (2018) [56]Cohort1,021 newbornsSmokingOverweight/obesityaOR 1.76 (1.14–2.71) p<0.01After adjustment, positive correlation between paternal smoking and offspring’s overweight/obesity at 5- and 9-years follow-up. Accordini et al (2018) [54]Cohort1,964 malesSmokingAsthmaRR 1.43 (1.01–2.01)Fathers’ smoking starting in early adolescence may independently increase asthma risk without aller- gies in offspring. Alati et al (2013) [10]Cohort7,062 malesAlcoholCognitive developmentKS2 score 0.27 (0.07–0.46)No strong evidence of intrauterine mechanisms link- ing paternal alcohol use with offspring cognitive development expressed in mean change of KS2 scoresa . Karalexi et al (2017) [57]Meta-analysis39 studiesAlcoholLeukemia (any type)OR 1.05 (0.91–1.22) p=0.931No association between paternal preconceptional alcohol consumption and risk of any type of leuke- mia cancer at 0–14 age. Acute lymphoblastic leukemiaOR 1.10 (0.93–1.30) p=0.361 Acute myeloid leukemiaOR 1.23 (0.83–1.82) p=0.489 Lindblad et al (2011) [59]Cohort 7,960 cases and 1,154,564 controls Addictive drugs disorderADHD Boys GirlsOR 3.7 (3.4–4.2) OR 4.6 (3.5–5.9)

Parental addiction to illicit drugs is associated with the highest OR for ADHD medication to treat the offspring. Fang et al (2018) [58]Cohort3,210 infantsOpioidsPremature deathaHR 4.79 (1.16–19.79) p<0.05The risk of premature death (0–6 years) increases 2.5–5.2 times in relation to the severity of paternal opioid use. Sallmén et al (2016) [62]Cohort11,863 cases and 23,720 controlsLead blood level (μmol/L) <0.5Schizophrenia, spectrum disorderaHR 0.97 (0.52–1.83)Paternal occupational exposure to lead does not increase risk for schizophrenia in the offspring no matter the metal blood level. 0.5–0.9aHR 1.25 (0.85–1.82) 1.0–1.4aHR 0.90 (0.54–1.49) ≥1.5 μmol/LaHR 1.38 (0.65–2.92) Nieuwenhuijsen et al (2013) [60]Meta-analysis3 studiesSolventsAnencephalyOR pooled 2.18 (1.52–3.11)Evidence for association between paternal solvents exposure and child NTDs or anencephaly. No caus- ative role of pesticides for hypospadias.NTDsOR pooled 1.86 (1.40–2.46) Spina bifidaOR pooled 1.59 (0.99–2.56) PesticidesHypospadiasRR 1.19 (1.00–1.41) Jørgensen et al (2014) [63]Cohort600,000 birthsPesticidesCryptorchidismaHR 1.04 (0.96–1.12)No increased risk of cryptorchidism after pesticide exposure.